Does the Definition of Borders of the Planum Temporale Influence the Results in Schizophrenia?

Several studies have used in vivo and postmortem methods to examine whether the lateralization of the planum temporale is disturbed in schizophrenia (1–6). Methodological differences between studies, such as the application of different definitions of planum temporale and planum temporale borders (1–5), may have contributed to the conflicting results. However, this has not been clarified. Barta et al. (6) found that different methods of surface rendering may yield different results with regard to planum temporale asymmetry and that the examination of planum temporale surface and planum temporale volume may produce different results in the same subjects. All volumetric planum temporale studies that detected asymmetry disturbances in patients with schizophrenia examined the total planum temporale (1, 4, 5), and studies that examined the anterior planum temporale failed to detect significant differences between patients and comparison subjects (2, 3).

To test the hypothesis that different definitions of planum temporale borders may influence the results with regard to a possible disturbance of planum temporale asymmetry in schizophrenia, we attempted to examine the anterior, posterior, and total planum temporale volumes simultaneously. Three frequently used definitions of the posterior border of the planum temporale proper were applied (4).

Method

Subjects

The study group comprised 30 right-handed, medicated male inpatients with schizophrenia (mean age=29.37 years, SD=7.97, range=18–47). Demographic information and clinical history were obtained by using a semistructured interview. The comparison group comprised 30 right-handed, healthy male comparison subjects who were recruited from the general population. Their mean age was 29.17 years (SD=8.03, range=18–47); they were matched to the patients in a one-to-one fashion for age (within ±3 years) and educational achievement. Handedness was evaluated by using the Edinburgh Inventory (7). Neither healthy subjects nor their first-degree relatives had a history of neurological or mental illness. Exclusion criteria for both groups were a history of serious neurological disorder, use of benzodiazepine or cortisol medication in the last 3 months, any mental disorder other than schizophrenia, and previous ECT. All patients and comparison subjects provided written informed consent after the procedures had been fully explained.

Measures

Magnetic resonance imaging (MRI) data were acquired by using coronal T₂-weighted and proton-density-weighted dual-echo sequences and three-dimensional magnetization-prepared rapid-gradient-echo sequences, described in detail elsewhere (8). Volumetric measurements of planum temporale volumes and total intracranial content were performed on the MRI data sets by using the software program BRAINS (9).

Details of the definition of the planum temporale borders and interrater and intrarater reliability are available elsewhere (8). Briefly, interactive measurements included a planum temporale total for each hemisphere, which was separated into anterior and posterior planum temporale by using three separate definitions.

Definition 1 was the region lateral to Heschl’s gyrus (planum temporale lateral to Heschl’s gyrus for the anterior definition and planum temporale posterior to Heschl’s gyrus for the posterior definition).

Definition 2 was the knife-cut method (the horizontal planum temporale anterior to the knife-cut for the anterior definition and the ascending planum temporale posterior to the knife-cut for the posterior definition).

Definition 3 was the bifurcation of Sylvian fissure (the region anterior to the branching point of the horizontal ramus of the Sylvian fissure into an ascending and descending ramus for the anterior definition and the region posterior to the branching point for the posterior definition).

The planum temporale total (anterior plus posterior) was identical in all definitions. Asymmetry coefficients were calculated according to the equation of Galaburda et al. (10).

Statistical Analysis

Possible confounding variables were assessed by using t tests for independent samples. Morphometric data were normally distributed. Relative volumes ([absolute region of interest volumes/intracranial content] × 100) were subjected to a two-factor analysis of variance (ANOVA) to assess the main and interaction effects of the between-subject factor diagnosis (patients, comparison subjects) and the within-subject factor (left hemisphere, right hemisphere). Unpaired Student’s t tests were used to assess the significance of differences in asymmetry coefficients.

Results

An overview of the results is given in Table 1. The groups did not differ significantly with respect to age (t=–0.97, df=58, p=0.92).

Regarding the total planum temporale, the two-factor ANOVA revealed no significant differences between the groups (F=0.69, df=1, 58, p=0.41). The total left planum temporale was significantly larger than the total right planum temporale (F=5.7, df=1, 58, p=0.02), without significant interaction between groups and hemispheres (F=0.26, df=1, 58, p=0.61).

Regarding the anterior planum temporale, the two-factor ANOVA found no significant volume differences between patients and healthy subjects for any of the three definitions (definition 1: F=1.0, df=1, 58, p=0.32; definition 2: F=1.1, df=1, 58, p=0.30; definition 3: F=1.3, df=1, 58, p=0.29).

Regarding the posterior planum temporale, the two-factor ANOVA found no significant volume differences between patients and healthy subjects for any of the three definitions (definition 1: F=0.02, df=1, 58, p=0.89; definition 2: F=0.05, df=1, 58, p=0.82; definition 3: F=0.19, df=1, 54, p=0.66).

No significant differences between men with and without schizophrenia emerged regarding asymmetry coefficients for any of the three definitions.

Discussion

To our knowledge, this investigation is the largest study to date on planum temporale volumes of male patients with schizophrenia. Our data suggest that the definition of planum temporale borders does not influence the results concerning possible disturbances of planum temporale asymmetry in schizophrenia. Therefore, results may reflect real morphological brain differences between our group of patients and those examined by others. There are four studies that examined the volume of the planum temporale in schizophrenia in vivo by MRI (2–5) and one that performed an in vitro examination of postmortem tissue (1).

Falkai et al. (1), who detected planum temporale asymmetry disturbances, examined the brains of chronically ill hospitalized patients. In contrast, patients in the Maudsley family study (2), which did not find significant differences in the asymmetry of the planum temporale between patients and control subjects, had a shorter illness duration than those examined by Kwon et al. (4) and Falkai et al. (1). These findings are consistent with the assumption that patients with a more chronic course of disease may show more brain abnormalities.

In conclusion, our results may support the neurodevelopmental hypothesis of more severe brain alterations in a subgroup of more severely ill patients with schizophrenia. Possible correlations between factors of chronicity, severity of the schizophrenic disturbances, and planum temporale measurements need to be evaluated in further studies.

TABLE 1

Enlarge table